make dtensor shared test util more generic

[vkuzo]

Summary:

Refactoring to make adding MX easier in next PR, no logic change. Specifically:

• move _test_lowp_mlp_tensor_parallelism_base to shared util file
• move config generation out of this function to the callsite
• instead of gating lowp allgather by rowwise flag, gate it by allgather_in_lowp flag

Test Plan:

./test/float8/test_dtensor.sh

Reviewers:

Subscribers:

Tasks:

Tags:

[vkuzo]

Stack from ghstack (oldest at bottom):

• enable to_mxfp8 cast for DTensor #2420
• -> make dtensor shared test util more generic #2416
• rename torchao.testing.float8 to torchao.testing.training #2415
• fix float8 training TP+SP integration tests #2414

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2416

• 📄 Preview Python docs built from this PR

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (2 Unrelated Failures)

As of commit 5eb2066 with merge base 4e3d019 ():

BROKEN TRUNK - The following jobs failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

• Run Regression Tests / test (CPU 2.7, linux.4xlarge, torch==2.7.0 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh) (trunk failure)
  test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_prepare_for_propagation_comparison
• Run Regression Tests / test (CUDA 2.7, linux.g5.12xlarge.nvidia.gpu, torch==2.7.0, cuda, 12.6) / linux-job (gh) (trunk failure)
  test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_prepare_for_propagation_comparison

This comment was automatically generated by Dr. CI and updates every 15 minutes.